The overarching goal of the proposed research is to investigate the association of prenatal exposure to endocrine disrupting compounds (EDCs: phthalates and bisphenol A) with DNA methylation patterns and childhood obesity using novel epigenomic approaches. The program provides a research phase (ROO) for Allan Just PhD, an environmental epidemiologist, to become an independent academic investigator in epigenetic epidemiology. Prior research has demonstrated that human prenatal exposures can alter the fetal epigenome, and animal models have shown that phthalates and bisphenol A impact methylation and obesity risk in the offspring. The proposed research uses methylation microarrays and new sequencing-based technologies to investigate methylation patterns in the umbilical cord blood of two cohorts of human children. An epigenome-wide methylation microarray (Illumina Human Methylation 450K BeadChip), already being conducted on umbilical cord blood DNA from 540 children, will measure methylation in 480,000 individual sites spanning almost all known genes. Statistical models will relate these methylation measures to prenatal EDC exposures and to measures of obesity at age 4. These models use new statistical approaches that account for multiple comparisons and reflect the dependence between methylation sites within the genome. Replication in an independent cohort (n=180), using the same design, assures that findings are consistent and generalizable. All samples and obesity measures are already collected or ongoing in the two cohorts (called PROGRESS and PRISM) under existing grants led by collaborators. In the second phase, a novel technique, targeted enrichment, which has not yet been applied in large-scale epigenetic epidemiologic studies, will be used in conjunction with bisulfite treatment and next generation sequencing; this approach will generate new high resolution measures of all methylation sites in identified regions of interest. This research should provide greater power to detect associations between EDC exposures, methylation patterns, and obesity in children. This approach may also contribute to future epigenetic epidemiological studies by comparing the